Академический Документы
Профессиональный Документы
Культура Документы
Received July 29th, 2011; revised November 11th, 2011; accepted December 15th, 2011
ABSTRACT
Crossdocking is defined as an operational strategy that moves items through consolidation centers or cross docks with-
out putting them into storage. As the need to move inventory faster increases, more logistics managers are turning to
crossdocking but the ability to execute such strategy well depends on good planning, dynamic scheduling and coordina-
tion. This paper introduces our research and development work on cross docking solution in three aspects: optimized
planning on container grouping, clustering, sequencing and allocating containers to docks; real-time scheduling handles
the dynamics of container arrivals and actual pallet transfers; and cross-docking coordination conducts real-time task
assignment/sequencing and resource management to deal with dynamic changes.
design, network design, and crossdocking operations model of a generic cross-docking facility [14] to examine
planning and scheduling. the operational risks associated with individual cross-
Rohrer [4] discussed modeling methods and issues as docking facilities within a company’s distribution net-
they are applied to crossdocking systems. He described work under a dynamic environment.
how simulation helps to ensure success in crossdocking In all crossdocking situations, the timing of delivery
system design by determining optimal hardware con- and pickup is crucial to effective operations. The per-
figuration and software control. Apt and Viswanathan [5] formance of crossdocking depends on good planning and
addressed a framework for understanding and designing scheduling. Li et al. [15] considered short term schedul-
crossdocking systems and discussed techniques that can ing of material handling inside the terminal for a given
improve the overall efficiencies of logistics and distribu- truck schedule. They model the jobs and resources as a
tion networks. Napolitano [6] classified the various types machine scheduling problem and present a meta-heuristic
of crossdocking operations. These include manufacturing, for its solution. Chen, Guo and Lim [10] studied cross-
distributor, transportation, retail and opportunistic cross- docking scheduling where time windows for deliveries
docking. Napolitano also describes crossdocking as the and pickups are considered. They also considered cross-
“JIT in the distribution arena”. Manufacturing cross- dock-handling costs which are use to penalize delays. Yu
docking constitutes the receiving and consolidating of and Egbelu [16] studied the scheduling issue of inbound
inbound supplies where a manufacturer can use a ware- and outbound trucks in crossdocking systems with tem-
house to receive supplies of parts for meeting demands porary storage. They try to find the scheduling sequence
ascertained from an MRP. In retail crossdocking, retail- for both inbound and outbound trucks to minimize total
ers receive products from multiple vendors who use dis- operation time when a storage buffer to hold items tem-
tributors with multiple warehouses. In general, cross- porarily is located at the shipping stock. Chen and Lee
docks are complex, requiring a high degree of coordina- [17] develop polynomial approximation algorithm and
tion between suppliers, customers and distributors to branch-and-bound algorithm to minimize the makespan
create shipments based on anticipated supplies and de- for products going through a crossdocking facility.
mands [7]. McWilliams et al. [18] covered a specific truck schedul-
Donaldson et al. [8] studied a network of crossdocks ing problem at a parcel hub. A simulation-based sched-
for the US Postal Service where 148 Area Distribution uling approach with an embedded genetic algorithm is
Centers serve as crossdocks, each receiving, sorting, proposed.
packing and dispatching mail according to operating Based on our reviews, crossdocking planning and sched-
schedules. Each distribution center serves as an origin as uling problem is still not properly investigated. Although
well as destination node where schedules were driven by some research proposed models on truck scheduling and
mail delivery standards. Ratliff et al. [9] studied a load- dock assignment problems related to crossdocking, most
driven network, in which deliveries take place when there of the studies are in the research stage, especially the
are sufficient products waiting for transportation. They time for getting the solution is still too long and un-fea-
studied the North American automobile delivery systems sible to real scenarios. Also most of the solutions do not
to determine the ideal number and location of crossdocks consider handling of uncertainties in the scheduling and
in a network and how shipments flowed between them. coordination of crossdocking operations. There is still
In their study, a minimum inventory strategy was the key shortage of effective operations planning solution for
in attempting to minimize the number of vehicles at the crossdocking with considerations of vehicle arriving, con-
tainer allocation, and exchanging sequencing. There is
mixing center (crossdocks). Chen, Guo and Lim [10]
also shortage of real-time scheduling for handling dy-
studied crossdocking network scheduling where time
namics on container arriving, allocation and resource
windows for deliveries and pickups are considered. They
uncertainties. These issues are discussed in this paper.
also considered crossdock-handling costs which are use
to penalize delays.
3. Research Issues
Also there are quite amount of research work on
crossdocking facility design. Barthold and Gue [11] de- As discussed previously, operational performance in
termined the best shape for a crossdock by analyzing the crossdocking greatly depends on good planning, sched-
assignment of receiving and shipping docks. The staging uling and coordination. We studied technologies and
of products in a crossdock to avoid floor congestion and algorithms to provide optimized solution for cross-dock-
increase throughput has also been studied together with ing planning, scheduling and operations coordination:
the effects of different combinations of number of work- Optimized planning for container grouping, clustering,
ers in receiving and shipping on throughput [12]. A sequencing and allocation with build-in mathematic
simulated annealing procedure was used to construct ef- algorithm.
fective layout to reduce labour costs [13]. A simulation Real-time scheduling handles the dynamics of con-
group is represented by the yellow color in Figure 2. doors especially when door becomes a critical resources.
Clustering: Identification of the containers (vertices) Currently, most crossdocks rely on supervisors to
with strong connectivity relationships within the same manually create docking assignments, or use WMS or
group. By this, the appropriate sequence of clusters of TMS solutions to schedule appointments directly to spe-
containers to enter the crossdocking area could be cific docking bays by time block. These simplistic ap-
determined since a cluster with outputs to other clus- proaches often result in sub-optimal sorting assignments
ters should be allocated earlier. A cluster is repre- due to their inability to account for the complexity of
sented in green color in Figure 2. sorting operations, especially during peak flows. In addi-
Identify minimum door requirements for groups (clus- tion, they cannot accommodate dynamic changes that are
ters): A crossdock needs to cater for multiple custom- inevitable in execution, leading to inefficient dock utili-
ers, so multiple group of containers compete for lim- zation and pallet transfers.
ited door resources. This function is to calculate the Unlike manual or other less capable appointment sys-
minimum number of docks needed is for completing tems, our cross-docking planning and scheduling engines
the exchange of a group without re-admission of con- increase sorting capacity by employing advanced model-
tainers to the crossdock. ing to consider container relationships, sorting resources,
Optimized allocation of groups and containers to docks: and facility configuration. Container groups are allocated
this is to determine which container should be as- to the right areas considering priorities. Bin-packing al-
signed to which door with consideration of minimiz- gorithms are used for optimal container allocation (Fig-
ing movement distance in pallet exchanges (Figure ure 3). Bins (refer to doors) are occupied in the time di-
3). mension with the right height by the right container to
Facility data, resource data, and pallet transferring re- make exchanging of pallets possible. Containers with
quirements are defined in Excel sheet as input to the high exchanging relationship are allocated to neighbor
planning model. An algorithm is also designed to identify docks to reduce exchanging cost and time. The plan en-
the minimum door requirements for groups and clusters sures that containers are admitted to cross-dock and pal-
which are very useful in the assignment of containers to lets are transferred in a smooth and continuous way.
in the crossdock, in the outside waiting area, or just leave There are three main modules in the coordination layer.
without get the input from the delayed container. The They are business application module, foundation mod-
system can redefine new groups (for the later container ule (workflow and database) and GUI module. The GUI
and the waiting container) and arrange time slot and module is for user’s interaction with the system including
docks for the new groups, or define virtual container to Main GUI for Control, Scenario Definition, Progress
be processed in the next day for the container which Monitoring and Control, Task Management, Resource
leave without taking the inputs from the delayed con- Management, Norms Management. The business appli-
tainers. cation module provides core business logics for process
The rescheduling module also includes the functions execution and support such as assignment implementa-
of grouping, clustering and sequencing and dock alloca- tion, process implementation, mode control for opera-
tion which are similar with the planning module. Cross- tional/auto-run, resource (forklift, dock & staging area)
docking scheduling and dynamic container admission controller, main application controller, and assignment
will combat peak and valley waves by balancing demand management (resource allocation, assignment and task
and resource availability and therefore smooth the work- fulfillment). Foundation module provides system level
load throughout based on the service availability of services to support fundamental object and information
cross-dock dock and forklift. management, database access and interface with work-
The admission control module is in-between of plan- flow engine.
ning layer and coordination layer, acts as the role for The coordination layer can be run standalone as well
real-time container admission to crossdock by taking as integrated with the scheduling layer. To run as a
schedule and real-time information from coordination standalone system, the data of containers and pallets can
layer as inputs, and generating real-time container ad- be loaded from an excel file and the information can be
mission plan based on input information. Admission lo- used to create internal objects. For integrated running of
gics are designed in such a way that the admission is the system, the containers will be admitted by admission
conducted based on existing schedule when there is a module in real-time based on the schedule that is created
container leaving the crossdock upon finishing its pallet dynamically according to real-time situation. In this way,
exchanging. However, re-scheduling will be performed if coordination layer is able to handle continuous container
container processing is very late or very earlier, dock is admission and was able to manipulate data in a real-time
broken-down, or container arriving is late etc. In the case manner. Information of containers and its pallets that are
of re-scheduling, containers will be admitted into cross- not admitted yet will be only used for planning and esti-
dock based on the new schedule. mation purpose in coordination.
An event-based integration framework has been de-
4.3. Coordination Platform Design and veloped for integrating the planning, scheduling and co-
Development ordination layers. It is a server-client integration mecha-
nism which can bring the advantages to the 3-layer
The purpose of crossdocking coordination is to maximize crossdocking system, such as portable and flexible con-
throughput and improve operations efficiency of cross- figuration and plug & play. An event server can has sev-
docking operations with resource constraints. A smooth eral event clients registering with it. Information can be
sorting workload ensures that the right resources are transferred among the clients when events are triggered.
available to avoid bottlenecks and maximize throughput. Data interface format has been defined in order for the 3
A cross-docking master scenario and related assump- layer to communicate and transfer data based on the
tions have been defined in the coordination layer. Proc- same protocol. It includes data interface format defined
esses for the handling of general containers and pallets for container-processing completion event, container ar-
have been modeled and confirmed. Task assignment and rives late event, dock service status change event, and
resource scheduling at the operational level match tasks container admission event. Event management mecha-
with resources. Task assignment logics and processes are nism has been implemented in the coordination layer and
designed into the system according to user requirements admission layer to apply the logics for events emitting,
defined. In addition, specifications for resources (con- receiving and handling.
tainers, cross-docking docks, forklifts, and staging areas)
and related control logics are included in the coordina- 5. Solution Validation and Advantages
tion module. Locations and distance norms are also de-
5.1. Prototype Testing
fined for container and forklift allocation purposes. Time
norms are defined for task control and assignment. GUI The input data for the prototype include container data,
requirements for different application modules have also container pallet exchange information, facility data (in-
been developed. clude distance between doors), and time norms informa-
tion. The container data represents the containers that well with dynamic changes. The advantages of the solu-
will go through the crossdock in a particular time period; tion include:
container pallet exchange information is about what pal- Maximize throughput and improve efficiency through
lets would be exchanged between containers; facility optimization and coordination such that containers ar-
location information include the relative distance be- rive at the right dock at the right time and pallets are
tween locations such as docks and staging area; and the transferred in the right sequence even in the face of
time norms information represents the time needed for a dynamics.
standard loading, unloading, or movement operations. Sorting based on optimized container allocation:
Input data for prototype testing and has been designed. unlike traditional systems, our cross-docking engines
We have debugged and tested the prototype system with increase sorting capacity by employing advanced
predefined input data for both different scenarios. The modelling to consider container relationships and fa-
simple scenario includes 7 pallets from 8 containers while cility configuration. Containers are admitted to cross-
the complex scenario is to test with the near real case for dock and pallets are transferred in a smooth and con-
390 pallets exchanging for 48 containers. Results ob- tinuous way.
tained from test runs show the system works by follow- Lower inventory cost and improved service levels:
ing expected logics and meet the targets. updating order quantities at the crossdock so that dis-
We have tested the planning and scheduling modules crepancies between what was ordered and what is
independently with multiple sets of data. We also have really needed could be reduced. This effectively re-
done integration testing of the planning and scheduling duces the safety stock, and also raises the service lev-
modules. The testing shows the communication between els.
scheduling and coordination are effective. The planning Reduce storage and material handling cost: Cross-
and scheduling result can be used to trigger operation and docking smoothly and speedily will not only lower
the status information could be feedback to rescheduling material handling costs but also contribute to lower
and it can trigger rescheduling if it is necessary according cost in the supply chain as storage is bypassed and
to the logic included. velocity is increased.
Consideration for dynamic effects and item priority:
5.2. Solution Advantages dynamic scheduling and admission will combat peak
and valley waves by balancing demand and resource
Logistics companies in many developing countries, espe-
availability.
cially in places like China, plan manually and are thus
The developed solution has the following benefits to
forced to organize themselves in simplified ways to han-
users:
dle real world complexities. They therefore suffer from
Suppliers and manufacturers would know exactly
lack of scale and know-how to exploit technology to
when the cargo should be sent to cross-docking center,
overcome this problem. The cross-docking development
and when the needed material would arrive. This will
in IDSC provides optimized solution for crossdocking
avoid sending vehicle too early or too late.
planning, scheduling and operation coordination.
Distribution and crossdocking center will enjoy easier
The developed solution has been presented to Singa-
and smoother crossdock operations with faster opti-
pore local and MNC logistics companies. They com-
mized planning and scheduling that maximizes the
mented that the solution was quite impressive and ex-
throughput of these facilities.
pressed high interest on the application of the solution.
Transportation service providers will see better plans
Discussion of potential application project is still in pro-
that coordinate their vehicle arrival times more accu-
gress. We also proposed to a China logistics company
rately and get their vehicles on the road again faster.
which has the requirement to speed up their forwarding
Recipients of the goods enjoy reduced inventory and
business for such customers as Wal-Mart. Since many
improved customer service levels as their inventory
countries around the world share the same challenges, the
can be replenished more frequently and stock-outs are
projected market size and sales for the technologies and
averted with higher velocity.
associated business model would be quite big. Another
major potential for optimal cross-docking planning and
real-time coordination is large-scale real-time military
6. Conclusions
logistics systems. The US military is leading the push in This paper introduces a solution developed for realistic
this area but Singapore’s growing sophistication in am- crossdocking operations. It includes three layers of mod-
munition logistics may also provide a boost. ules: planning layer for container arriving, grouping and
The solution aims to improve cross-docking operations allocation; scheduling layer for dynamic event handling;
not only with higher speed and lower cost, but also copes and coordination layer for execution control and resource